Generalized Linear Models for Insurance Rating

Generalized linear models have been in use for over thirty years, and there is no shortage of textbooks and scholarly articles on their underlying theory and application in solving any number of useful problems. Actuaries have for many years used GLMs to classify risks, but it is only relatively recently that levels of interest and rates of adoption have increased to the point where it now seems as though they are near-ubiquitous. GLMs are widely used in the personal lines insurance marketplace, especially in operations of meaningful scale. But as far as the authors are aware there is no single text written for the practicing actuary that serves as a definitive reference for the use of GLMs in classification ratemaking. This monograph aims to bridge that gap. Our ultimate goal is to give the knowledgeable reader all of the additional tools he or she needs to build a market-ready classification plan from raw premium and loss data.

The target audience of this monograph is a credentialed or very nearly credentialed actuary working in the field of property/casualty or general insurance (for example, in the United States, a member or soon-to-be member of the Casualty Actuarial Society). It is assumed that the reader will be familiar with the material covered in the earlier exams of the CAS syllabus, including all of the Actuarial Standards of Practice and the ratemaking material covered in depth in Werner and Modlin’s Basic Ratemaking (2010) (or their international equivalents, for readers outside the United States). Prior knowledge of the mathematics underlying GLMs will make for faster reading but is not absolutely necessary. Familiarity with a programming language is not required to read the text, but will be necessary to implement models.